RWKV-PCSSC: Exploring RWKV Model for Point Cloud Semantic Scene Completion

TL;DR

This paper introduces RWKV-PCSSC, a lightweight and efficient neural network for semantic scene completion from point clouds, leveraging the RWKV mechanism to reduce complexity while achieving state-of-the-art results.

Contribution

We propose a novel lightweight network architecture, RWKV-PCSSC, that uses RWKV modules for efficient feature aggregation and restoration in point cloud scene completion.

Findings

Reduces parameter count by 4.18 times compared to state-of-the-art.

Improves memory efficiency by 1.37 times.

Achieves state-of-the-art performance on multiple indoor and outdoor datasets.

Abstract

Semantic Scene Completion (SSC) aims to generate a complete semantic scene from an incomplete input. Existing approaches often employ dense network architectures with a high parameter count, leading to increased model complexity and resource demands. To address these limitations, we propose RWKV-PCSSC, a lightweight point cloud semantic scene completion network inspired by the Receptance Weighted Key Value (RWKV) mechanism. Specifically, we introduce a RWKV Seed Generator (RWKV-SG) module that can aggregate features from a partial point cloud to produce a coarse point cloud with coarse features. Subsequently, the point-wise feature of the point cloud is progressively restored through multiple stages of the RWKV Point Deconvolution (RWKV-PD) modules. By leveraging a compact and efficient design, our method achieves a lightweight model representation. Experimental results demonstrate that RWKV-PCSSC reduces the parameter count by 4.18$\times$ and improves memory efficiency by 1.37$\times$ compared to state-of-the-art methods PointSSC. Furthermore, our network achieves state-of-the-art performance on established indoor (SSC-PC, NYUCAD-PC) and outdoor (PointSSC) scene dataset, as well as on our proposed datasets (NYUCAD-PC-V2, 3D-FRONT-PC).